6 Reading the consensus table
The consensus table bundles all key metadata you need: origin, species support, scores, and IDs for plotting and networks.
top_consensus <- cons %>%
dplyr::filter(significant_in_any) %>%
dplyr::arrange(dplyr::desc(consensus_score)) %>%
dplyr::select(
term_id, term_name, ontology,
num_species_gprof_bg, num_species_gprof_nobg,
consensus_score, consensus_score_all,
origin, source_origin
) %>%
head(25)
top_consensus
#> # A tibble: 25 × 9
#> term_id term_name ontology num_species_gprof_bg num_species_gprof_nobg
#> <chr> <chr> <chr> <dbl> <dbl>
#> 1 KEGG:00000 KEGG root te… KEGG 1 0
#> 2 GO:0009987 cellular pro… BP 2 3
#> 3 GO:0003674 molecular_fu… MF 2 0
#> 4 GO:0008150 biological_p… BP 2 0
#> 5 GO:0110165 cellular ana… CC 2 3
#> 6 GO:0005575 cellular_com… CC 2 0
#> 7 KEGG:03040 Spliceosome KEGG 1 3
#> 8 GO:0016020 membrane CC 1 0
#> 9 GO:0005737 cytoplasm CC 1 3
#> 10 GO:0005622 intracellula… CC 1 3
#> # ℹ 15 more rows
#> # ℹ 4 more variables: consensus_score <dbl>, consensus_score_all <dbl>,
#> # origin <chr>, source_origin <chr>Typical columns worth focusing on:
term_id,term_name,ontology– GO identifier, label, and ontology (BP/MF/CC).origin– high-level origin categories, e.g., “GO terms – Consensus (with BG)”, “GO terms – GOseq only”, “GO terms – g:Profiler only (no BG)”.source_origin– fine-grained tag summarizing the combination of sources, e.g.,"GOseq+g:Profiler_BG","g:Profiler_BG+g:Profiler_noBG".num_species_gprof_bg,num_species_gprof_nobg– number of species supporting the term via g:Profiler (with and without background).consensus_score,consensus_score_all– strict vs. exploratory composite scores.significant_in_any–TRUEif any of GOseq, g:Profiler with BG, or g:Profiler no-BG report the term as significant.
Figure 1 illustrates how these scores and annotations translate into a ranked list of biological processes in the demo dataset.
Figure 1. Exploratory EchoGO lollipop plot for BP terms in
the demo dataset.
Top exploratory BP GO/KEGG terms are ranked by the EchoGO exploratory
consensus score and plotted as a lollipop chart. Each point corresponds
to a GO or KEGG term, and the x-axis shows the ranking score used by
EchoGO to prioritize terms (higher = stronger, more consistent signal
across tools and species).
Exploratory EchoGO lollipop plot for BP
terms in the demo dataset.
Reading tip: For high-confidence biological interpretation, prioritize terms that:
- Have high
consensus_score, - Are supported by GOseq and g:Profiler with background,
- Have multi-species support
(
num_species_gprof_bg ≥ 2), - Form coherent clusters in RRvGO and network outputs.
